Sat-MVS: Multi-View Stereo Dense Matching Network for Satellite Images

Official Implementation of ICCV2020: Rational Polynomial Camera Model Warping for Deep Learning Based Satellite Multi-View Stereo Matching

Requirements

For more details, please refer to environment.yaml. And You can simply import this environment from the yaml file via conda:

conda env create -f environment.yaml

conda activate satmvs

Some packages are list here:

| package | version | | -------------- | -------- | | gdal | 3.3.1 | | matplotlib | 3.4.3 | | numpy | 1.12.5 | | tensorboardx | 2.5 | | pytorch | 1.4.0 | | torchvision | 0.5.0 | | numpy-groupies | 0.9.14 | | opencv-python | 4.5.5.62 |

Data Preparation

See WHU_TLC/readme.md for more details. And rename the "open_dataset" to "open_dataset_rpc".

Pretrain models

You can download the models at: https://pan.baidu.com/s/1_z_o1ozWryIt7J05l-Rp_w?pwd=xo2p code: xo2p

Train

Train on WHU-TLC dataset using RPC warping:

python train.py --mode="train" --model="red" --geo_model="rpc" --dataset_root=[Your dataset root] --batch_size=1 --min_interval=[GSD(resolution of the image)] --gpu_id="0"

Train on WHU-TLC dataset using homography warping:

python train.py --mode="train" --model="red" --geo_model="pinhole" --dataset_root=[Your dataset root] --batch_size=1 --min_interval=[GSD(resolution of the image)] --gpu_id="0"

Predict

If you want to predict your own dataset, you need to If you want to predict on your own dataset, you need to first organize your dataset into a folder similar to the WHU-TLC dataset. And then run:

python predict.py --model="red" --geo_model="rpc" --dataset_root=[Your dataset] --loadckpt=[A checkpoint]

Citation

If you find this work helpful, please cite our work: @InProceedings{Sat_MVS, author = {Gao, Jian and Liu, Jin and Ji, Shunping},> title = {Rational Polynomial Camera Model Warping for Deep Learning Based Satellite Multi-View Stereo Matching}, booktitle = {Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV)}, month = {October}, year = {2021}, pages = {6148-6157} }

Acknowledgements

Thanks to the authors of UCS-Net, Cas-MVSNet, and VisSat (adapted COLMAP) for open sourcing their fantastic projects. You may want to visit these projects at:

https://github.com/touristCheng/UCSNet

https://github.com/alibaba/cascade-stereo

https://github.com/Kai-46/VisSatSatelliteStereo